Equilibrium and Thermodynamics on Arsenic(III) Sorption Reaction in

Mar 17, 2010 - Tina Basu, Kaushik Gupta and Uday Chand Ghosh*. Department of Chemistry, Presidency College, 86/1 College Street, Kolkata 700 073, Indi...
1 downloads 0 Views 460KB Size
Download PDF
J. Chem. Eng. Data 2010, 55, 2039–2047

2039

Equilibrium and Thermodynamics on Arsenic(III) Sorption Reaction in the Presence of Background Ions Occurring in Groundwater with Nanoparticle Agglomerates of Hydrous Iron(III) + Chromium(III) Mixed Oxide† Tina Basu, Kaushik Gupta, and Uday Chand Ghosh* Department of Chemistry, Presidency College, 86/1 College Street, Kolkata 700 073, India

Agglomerated nanoparticles [(40 to 50) nm] of synthetic iron(III) + chromium(III) mixed oxide (NHICO) were investigated for the equilibrium and thermodynamics of arsenic(III) removal, respectively, at pH 7.0 (( 0.2) and temperature (T ( 1.6) ) (288 and 303) K from solutions with background ions that coexist in groundwater. Langmuir and Redlich-Peterson isotherm models described the equilibrium data better than Freundlich model. The Langmuir capacity (θ0 · 102, mmol · kg-1) values estimated, 10.404 and 10.686 in the absence of ions at studied temperatures, were higher than that in the presence of ions (except for HCO3-) at background. The enthalpy (∆H0) and entropy (∆S0) changes were positive, indicating endothermic reactions driven by entropy increase at the solid-liquid interface. Negative values of free energy change (∆G0) indicated that the reactions were spontaneous, which increased with increasing temperature on reactions. The treatment of groundwater (arsenic concentration: 1.347 µmol · dm-3) by the NHICO packed fixed-bed column (height: 6.0 cm, i.d.: 1.0 cm, bed volume: 4.71 cm3) yielded 4.7 dm3 (∼1000 BV) of water with an arsenic concentration of e 0.133 µmol · dm-3. The used bed was regenerated up to a level 90 (( 1.5) % of its initial capacity with 0.350 dm3 of optimized 0.25 M NaOH solution.

Introduction Arsenic exists in the natural environment in four different oxidation states: As(V), As(III), As(0), and As(-III). The mobility and toxicity of arsenic are determined by its oxidation states;1 thus, the behavior of arsenic species will change depending on biotic and abiotic conditions in water. The occurrence of arsenic in groundwater much exceeding the tolerance limit (10 µg · dm-3/0.133 µmol · dm-3) is a global problem2 and poses an ever-increasing degree of health hazard. The Bengal Delta Basin (West Bengal in India and Bangladesh) has become infested with this menace, and in some pockets of this region it has assumed a life-threatening proportion, causing deaths of a good number of inhabitants. An accumulation of excess arsenic in groundwater in this delta region is presumably due to the microbial reduction of geogenic arsenopyrites and iron oxyhydroxide with adsorbed arsenic in an anoxic environment.3,4 The aquifers thus become rich in this reduced As(III) along with Fe(II). The ratios of As(III)/As(Total) at a depth of (30 to 40) m reported in these aquifers are in the range of 0.6 to 0.9,4 which is a matter of great concern since the toxicity and mobility of arsenic(III) are greater than those of arsenic(V). The high toxicity of arsenic(III) is due to the high combining affinity with the thiol (-SH) part of the protein due to a soft-soft acid-base reaction. Remembering the adverse health impact of arsenic toxicity for long-term drinking of groundwater contaminated with high arsenic, the researchers had paid attention for last two decades in developing methods for reducing arsenic level below or equal to the permissible value (10 µg · dm-3/0.133 µmol · dm-3). Consequently, several methods such as oxidation-precipitation, † Part of the “Josef M. G. Barthel Festschrift”. * Corresponding author: Tel./Fax: +91-33-2241-3893. E-mail: ucghosh@ yahoo.co.in.

coagulation/electrocoagulation/precipitation, membrane filtration, surface sorption, and ion exchange, and so forth, were reported.2,5,6 However, the surface sorption has been found to be an alternative option for the treatment of arsenic-rich groundwater and has been well-accepted by the rural people of under-developed countries like India and Bangladesh for its simplicity of operation, requirement of less space, and low recurring cost. Numerous sorbent materials,5,6 namely, activated carbon, agricultural products and byproduct, biomasses and metal oxides, or metal ion loaded biomaterials were tested for the treatment of arsenic-rich (> 0.133 µmol · dm-3) ground/ wastewater and industrial effluents. It has been found that different mineralogical forms of iron(III)/aluminum(III) oxide and hydroxide7-22 in the bulk phase were used for arsenic sorption from aqueous solutions. Arsenic sorption or removal using some synthetic polyvalent metal oxides had also been investigated22-27 in our laboratory. Zhang28 demonstrated an overview on the use of nanoscale iron particles for environmental remediation for large surface areas and in situ reactivity. Consequently, the sorption behavior of arsenic from aqueous solutions had been reported using nanoscale zerovalent iron29,30 and nanocrystalline titanium oxide.31,32 On the basis of these studies, some filters for the treatment of arsenic-rich groundwater are developed and marketed by different companies. It had been reported33 that the assessment performances of the filters at the field were not up to the level of specification given by the manufacturers. It is presumably due to the adverse influence of dissolved ions like sulfate, phosphate, chloride, calcium, magnesium, and so forth that those co-occur with arsenic in groundwater. Therefore, we have aimed for the sorption of arsenic by iron(III)-based sorbent materials to be investigated for removal from water in the presence of some background ions, which occur with arsenic in groundwater.

10.1021/je901010x  2010 American Chemical Society Published on Web 03/17/2010

2040

Journal of Chemical & Engineering Data, Vol. 55, No. 5, 2010

Thus, this manuscript reports herein the results of equilibrium and thermodynamics studies for the sorption reaction of arsenic(III) with nanoparticle [(40 to 50) nm] agglomerates of synthetic hydrous iron(III) + chromium(III) mixed oxide (NHICO) in the presence of some background ions occurring in groundwater with the column treatment of high arsenic groundwater by a NHICO packed fixed bed.

Materials and Methods Chemicals. All chemicals used were of guaranteed reagent (G.R.) grade (E. Merck, India) except for arsenic(III) oxide (As2O3; 99.99 %, Analar Reagent, BDH, England), iron(III) chloride (FeCl3; 98 %, laboratory reagent, E. Merck, India), sodium hydroxide (NaOH; 97 %, laboratory reagent, SD Fine Chemicals, India), and chromium(III) chloride (CrCl3; 93 %, laboratory reagent, Loba Chemie, India). Arsenic(III) Solution. A standard stock arsenic(III) solution (13.33 mmol · dm-3) was prepared by dissolving 1.76 mmol of arsenic(III) oxide in 0.1 dm3 of arsenic-free 0.04 mass fraction sodium hydroxide solution, which was diluted with doubledistilled water to 1.0 dm3 into a volumetric flask. A measured volume of this stock solution was diluted with 0.002 volume fraction of HCl in water for a desired level of arsenic(III) concentration in solution prior to use in experiments. The stock was prepared fresh after every 15 days and frozen to prevent oxidation. Arsenic Analysis. Concentrations of arsenic in sample solutions were analyzed using a UV-vis spectrophotometer (Hitachi, model U-3210) and atomic absorption spectrophotometer (Perkin-Elmer, model Analyst 200) by the methods described in the Standard Methods for the Examination of Water and Wastewater.34 Synthesis of NHICO. NHICO was synthesized by the method of chemical precipitation. Here, FeCl3 (0.1 M in 0.1 M HCl) and CrCl3 (0.1 M in 0.1 M HCl) were mixed together in 1:1 volume ratios and warmed at 353 (( 8) K. To the hot, wellstirred (speed: ∼ 300 rpm) mixed solution, aqueous NH4OH (1:1, volume fraction) was added carefully until the pH reached between 5.5 and 6.7. The brown jell-like precipitates that appeared were aged with liquid for 4 days, filtered, and washed three times with distilled water. It was dried inside an air oven at (333 to 343) K. The dried hot mass when treated with cold water was broken to the fine agglomerates. The grain size of agglomerates ranging from (140 to 290) µm were sieved out and characterized for the nanostructure. These grains were heattreated at (323, 373, 473, 523, 573, 623, 673, 723, and 773) K for an hour in a muffle furnace and desiccated at room temperature (303 ( 1.6 K), which was used for the experiments. Batch Experiments. To evaluate optimum temperature for the heat treatment on NHICO, 0.05 g of solid heat-treated at different temperatures was added with 0.05 dm3 of arsenic(III) solution [concentration, CI ) (0.10 and 0.13) mmol · dm-3 and pH ) 6.4 ( 0.2] into the polyethylene tetraphthalate (PET) bottles (capacity: 0.25 dm3). Reaction mixtures in PET bottles were agitated (speed: 350 ( 5 rpm) using a thermostat shaker at temperature (T) ) 303 ( 1.6 K for 2.0 h and filtered. Arsenic concentration (Ce, mmol · dm-3) was analyzed in each filtered solution. For the effect of initial solution pH (pHi) on the removal of arsenic(III) from aqueous solutions, 0.05 g of NHICO (heattreated at 623 K) was mixed with 0.05 dm3 of arsenic(III) solutions of CI ) (0.10 and 0.13) mmol · dm-3 adjusted separately at pHi ) 2.0, 3.0. 4.0, 5.0, 6.0, 7.0, 8.0, 9.0, and 10.0 into the PET bottles (capacity: 0.25 dm3) and agitated

(speed: 350 ( 5 rpm) for a duration of 2.0 h at T ) 303 ( 1.6 K. The equilibrium solution pH (pHf) value of each mixture was recorded immediately after 2 h of agitation by immersing the pH-meter electrode (model LI-127, ELICO India) and filtered to separate the solution from the solid particles. Each filtrate was analyzed for the arsenic concentration (Ce, mmol · dm-3) remaining at equilibrium. Batch procedure, which was adopted for the determination of the equilibrium time of arsenic(III) removal reaction with NHICO (heat-treated at 623 K) from the solution of CI ) (0.067 and 0.172) mmol · dm-3 including the background ion [SO42(2.08), Cl- (11.27), PO43- (0.53), HCO3- (4.92), Mg2+ (4.10), Ca2+ (2.5), SiO32- (0.64)] at pHi ) 7.0 (( 0.2) and T ) 303 (( 1.6) K was described in the Supporting Information (SI). For equilibrium isotherm, experiments were set as described in the first paragraph in this subsection. Here, the arsenic(III) solution of CI ranging from (0.13 to 3.33) mmol · dm-3 including a background ion [SO42- (2.08), Cl- (11.27), PO43- (0.53), HCO3- (4.92), Mg2+ (4.10), Ca2+ (2.5), SiO32- (0.64)] at pHi 7.0 (( 0.2) was mixed separately with NHICO (heat-treated at 623 K). Values noted in parentheses against each background ion were the CI (mmol · dm-3) of that ion added with arsenic(III) solution. A major change of pH during agitation, if any, in each set was checked at t ≈ (45 and 90) min from the time of start (t ) 0) and adjusted (if required) by 0.1 M HCl or 0.1 M NaOH. Reaction mixtures were filtered immediately after 2.0 h of agitation, and the filtrates were analyzed for arsenic concentration (Ce, mmol · dm-3) remaining at equilibrium. Calculation of Sorption Capacity. Arsenic(III) sorption capacity (qe, mmol · kg-1) at the equilibrium of NHICO was calculated from the following mass balance relation (eq 1):

qe ) [CI - Ce]·(V/W)·103

(1) -3

where CI and Ce are the concentrations (mmol · dm ) of arsenic in solution at t ) 0 and equilibrium, respectively; V and W are the volume (dm3) of solution and mass (g) of sorbent taken for the experiments. The multiplication factor 103 converted the qe values in terms of mmol · kg-1. Fixed-Bed Column Experiment for Arsenic RemoWal from Groundwater. A fixed-bed column was made by uniform packing of NHICO into a glass tube (i.d. 1.0 cm) up to a height of 6.0 cm (bed volume, BV: 4.71 cm3). Groundwater sample that was collected from a tube well [depth: (40 to 45) m] from Kalinarayan pore at the Nadia district in West Bengal (India) was analyzed for arsenic (0.101 mg · dm-3/1.347 µmol · dm-3). Some analyzed parameters for water quality (mmol · dm-3, except pH) were pH (7.5), Fe2+ (2.686 · 10-3), F- (0.034), and HCO3- (11.061), and hardness (2.92 as CaCO3), Mg2+ (0.547) and Ca2+ (2.38). This water sample was passed through the NHICO packed bed for filtration (flow rate: 0.06 dm3 · h-1). The filtered water collected by 0.1 dm3 fractions was analyzed for arsenic concentration. The arsenic-loaded NHICO bed in the column was regenerated by passing 0.25 M NaOH solution with an effluent rate of 0.06 dm3 · h-1. Effluents were collected by a fraction of 0.05 dm3 and analyzed for the arsenic released.

Results and Discussion Physicochemical Characterization of NHICO. An analysis of the X-ray diffraction (XRD) pattern (figure omitted) of NHICO suggested that the material was amorphous. Absorbance bands (spectra a to d in Figure S1 of SI) at wavenumbers (ν, cm-1) ranging from 3650 to 3120 and 1630 to 1400 were the stretching and bending modes of O-H bonds of lattice water

Journal of Chemical & Engineering Data, Vol. 55, No. 5, 2010 2041

and the hydroxyl groups, respectively. An absorption band appearing at ν ∼ 521 cm-1 in the spectrum of NHICO (marked as a) was absent in the spectra of the solid-solid mixture of iron(III) and chromium(III) oxides (marked as b), iron(III) oxide (marked as c), and chromum(III) oxide (marked as d). This indicated that the oxo/hydroxo bridge bond developed between iron(III) and chromium(III) in NHICO. Thus, it was found that the synthetic material was a hydrated mixed oxide. The thermogravimetric (TG) spectrum (Figure S2 of SI) showed a total weight loss by 24.75 % at temperatures between (80 and 800) °C which confirmed the hydrous nature of NHICO. Transmission electron microscopic (TEM) images (Figure S3 of SI) of NHICO confirmed the existence of nanoparticles [dimensions: (40 to 50) nm] in the agglomerates. Scanning electron microscopic (SEM) images (Figure S4 of SI) showed that the surface morphology of that material was irregular. The specific surface area, pHzpc, and bulk density of NHICO, respectively, were estimated to be 601.6 (( 5.0) m2 · g-1 and [6.4 (( 0.2) and 2.20 (( 0.02)] g · cm-3 (SI). Taking the weight loss data from TG analysis at > 80 °C and chemical analyses by the wet method of dissolved NHICO in HCl for iron and chromium suggested that the percentage composition of NHICO was Fe2O3 (36.54 ( 0.45), Cr2O3 (35.84 ( 0.60), and H2O (24.75 ( 0.35). Effect of Heat Treatment Temperature. Arsenic(III) sorption capacity (qe, mmol · kg-1) of heat-treated NHICO from aqueous solutions [CI of arsenic(III) ) (0.101 and 0.13) mmol · dm-3] at pH 6.4 (( 0.2) versus heat treatment temperature (Figure S5), described in the SI, showed that the value of qe increased with increasing heat treatment temperature on NHICO from (323 to 623) K and decreased with that from (623 to 773) K. Thus, heat-treated NHICO at 623 K had the highest arsenic(III) removal capacity from the aqueous solutions, and that NHICO was used for the remaining experiments. Effect of pH. The effect of initial solution pH (pHi) on arsenic(III) removal capacity from aqueous solutions (Figure S6 of the SI) is discussed in detail in the SI. It was found that the NHICO (heat-treated at 623 K) had the highest arsenic(III) removal capacity from the aqueous solutions at pHi ) 7.0 (( 0.2) and T ) 303 (( 1.6 K). Equilibrium Modeling. Figures 1 and 2 demonstrate the values (as points) of sorption capacity at equilibrium (qe, mmol · kg-1) against concentration at equilibrium (Ce, mmol · dm-3) of arsenic(III) for the sorption reaction with NHICO at T ) [288 (( 1.6) and 303 (( 1.6)] K, respectively. To understand the sorption mechanism, equilibrium data shown as points were analyzed by the isotherm model equations,35 namely, the Langmuir (eq 2), Freundlich (eq 3), and RedlichPeterson (RP) (eq 4) with nonlinear regression fit method on origin spread sheet.

Langmuir equation: qe ) (θ0KaCe)/(1 + KaCe)

(2)

Freundlich equation: qe ) KFCe1/n

(3)

RP equation: qe ) (ACe)/(1 + BCeg)

(4)

where θ0 and Ka are the Langmuir monolayer adsorption capacity (mmol · kg-1) and equilibrium constant (dm3 · g-1) and KF and n are the Freundlich constants related to the adsorption capacity (mmol1-1/n · kg-1 · dm-3/n) and intensity (dimensionless), respectively. Significances of qe and Ce are given above. The value of n is related to the affinity or binding strength of the solute with the solid surface. A (dm3 · kg-1) and B (dm3 · mmol-1) are the RP constants related to the adsorption equilibrium constant. The value of g, a dimensionless RP constant, should

Figure 1. Plots of equilibrium sorption capacity (qe) versus equilibrium concentration (Ce) of the arsenic(III) sorption reaction with NHICO at T ) 288 (( 1.6) K and pHi ) 7.0 (( 0.2) in the presence of some background ions and nonlinear fits of the data with (a) Langmuir, (b) Freundlich, and (c) RP isotherm equations. 9, no ion; b, SO42-; 2, Cl-; 1, PO43-; [, HCO3-; left-pointing triangle, Mg2+; right-pointing triangle, Ca2+; and `, SiO32-.

vary from 0 to 1.0. When g ) 0, the RP eq 4 converts to Henry law, and when g ) 1, that equation converts to Langmuir eq 2. Figures 1a-c and 2a-c show the nonlinear fits of the equilibrium data with the Langmuir (eq 2), Freundlich (eq 3), and RP (eq 4) equations. The estimated isotherm parameters for eqs 2 to 4 are shown in Tables 1 and 2 with the regression coefficient (R2) and statistical error χ2 values. On the basis of the values of χ2 and R2 (Tables 1 and 2), it was found that the present data described the Langmuir (eq 2) and the RP (eq 4) model isotherms very well (R2 g 0.99). The goodness-of-fit of

2042

Journal of Chemical & Engineering Data, Vol. 55, No. 5, 2010

the adsorption sites of the NHICO surface were homogeneous and accessed equally by the arsenic species. Values of the dimensionless Freundlich constant (n) laid between 1.73 and 2.29, indicating a high affinity of arsenic(III) for the solid surface. Moreover, the values of g of the RP model were close to 1.0 (0.78 < g < 1.20), which helped us to understand that the Langmuir (eq 2) model is a better fit of the present equilibrium data. This confirmed that the active sites available on the NHICO surface are equally accessible to the solute particles. In contrast to the isotherm modeling of equilibrium data by Kuriakose et al.19 and Zhang and Itoh,36 the modeling of present data was found to be almost similar with the others.8,9,12,18 Values for θ0 ( · 102) obtained were (10.4 and 10.7) mmol · kg-1 (Tables 1 and 2) for the arsenic(III) sorption reaction with NHICO in the absence of background ions, respectively, at T (( 1.6) ) (288 and 303) K, which were higher than those obtained for the reaction in the presence of background ions (except for HCO3-). Thus, the background ions (except for HCO3-) had an adverse effect on the removal of arsenic(III) by NHICO. The obtained values of θ0 ( · 102), (11.1 and 11.7) mmol · kg-1, respectively, at T (( 1.6) ) (288 and 303) K (Tables 1 and 2) in the presence of HCO3- in the background on the arsenic(III) sorption reaction with NHICO, were higher than those obtained in the absence of background ions. It indicated that the HCO3- had a positive effect on the equilibrium of the arsenic(III) removal reaction. This is probably for the faster sorption of bicarbonate by NHICO (log K ) 22.33 ( 0.01 for bicarbonate sorption on goethite)37 than As(OH)3 (log K ) 4.91, log K ) 7.26 for As(OH)3 sorption on goethite),38 which converted the negative solid surface and As(OH)3 sorbed strongly thereon via an intermolecular type of hydrogen bond formation. The separation factor,39 RL (a dimensionless parameter), which could predict the reaction feasibility had been expressed as,

RL ) (1 + KaCI)-1

Figure 2. Plots of equilibrium sorption capacity (qe) versus equilibrium concentration (Ce) of the arsenic(III) sorption reaction with NHICO at T ) 303 (( 1.6) K and pHi ) 7.0 (( 0.2) in the presence of some background ions and nonlinear fits of the data with (a) Langmuir, (b) Freundlich, and (c) RP isotherm equations. 9, no ion; b, SO42-; 2, Cl-; 1, PO43-; [, HCO3-; left-pointing triangle, Mg2+; right-pointing triangle, Ca2+; and `, SiO32-.

The significance of Ka and CI has been given elsewhere in this manuscript. When the value of RL is equal to (i) 0.0 to 1.0, the reaction is favorable, (ii) 0.0, the reaction is irreversible, and (iii) > 1.0, the reaction is unfavorable. Inserting the values of Ka (Tables 1 and 2) and CI of arsenic(III) [) (0.13 to 3.33) mmol · dm-3] into the depicted relation (eq 5), the RL values obtained were in the range of 0.0 to 1.0. This suggested that the arsenic(III) sorption reaction with NHICO, even in presence of background ions, was a favorable process at the conditions used for the experiments. Mean Sorption Energy. Mean sorption energy (Ea, kJ · mol-1), which helps to predict the nature of a reaction, is defined as the free energy of transfer of one mole of solute from infinity (in solution) to the surface of the sorbent. The value of Ea is related with the Dubinin-Radushkevich40 (DR) constant, BD (mol2 · kJ-2) as Ea ) (2BD)-0.5, which could be computed from the analysis of equilibrium data of a sorption reaction with the DR isotherm equation,40

ln qe ) ln qm - BDε2 the data with either the Langmuir (eq 2) or the RP (eq 4) model was comparatively better than that with the Freundlich model (eq 3; R2 ) 0.97 to 0.99). Assumptions based on which the Langmuir model (eq 2) is developed are (i) the monolayer adsorption of adsorbate on the adsorbent surface and (ii) the homogeneous energetic distribution of the active binding sites that are equally accessible to the solute. Good fits of the equilibrium data with the Langmuir model (eq 2) indicated that

(5)

(6) -1

where qm is the saturation sorption capacity (mmol · kg ). The Polanyi potential (ε) expressed as ε ) RT ln{1 + (Ce)-1}, where R is the molar gas constant and T, the absolute temperature. The value of BD was calculated from the slope of each linear plot of ln qe against ε2 (Figure S7 of SI) and evaluated the value for Ea. The values obtained are shown in Table 4. Values of Ea for the reaction of arsenic(III) sorption with NHICO in the

Journal of Chemical & Engineering Data, Vol. 55, No. 5, 2010 2043 Table 1. Isotherm Parameters Estimated by the Nonlinear Method of Analysis of Equilibrium Arsenic(III) Sorption Data on NHICO at pH ) 7.0 (( 0.2) and T ) 288 (( 1.6) K Langmuir isotherm parameters θ0 ( · 10 ) 2

As(III) + ion added 3-

PO4 SiO32Ca2+ ClSO42Mg2+ no ion HCO3-

Freundlich isotherm parameters Ka

KF ( · 102) n

R2

χ2 ( · 102)

A ( · 102)

B

g

2.018 2.291 2.183 1.928 1.727 1.982 1.793 2.075

0.998 0.993 0.995 0.993 0.996 0.997 0.998 0.991

0.751 3.968 3.432 5.905 2.45 2.798 1.65 13.868

9.821 15.191 16.826 10.884 8.21 16.318 11.679 24.021

1.758 2.569 2.37 1.217 1.002 1.872 1.082 2.279

0.883 0.901 0.912 1.026 0.933 0.904 1.037 0.938

R2

χ2 ( · 102)

mmol · kg-1

dm3 · mmol-1

R2

χ2 ( · 102)

(mmol · kg1-) · (dm3 · mg-1)1/n

0.998 0.992 0.995 0.993 0.996 0.997 0.998 0.99

0.936 3.685 3.282 4.94 2.107 2.715 1.419 11.775

6.427 6.58 7.766 8.697 8.883 9.614 10.404 11.128

1.299 1.944 1.886 1.286 0.869 1.495 1.159 1.973

0.983 0.975 0.975 0.971 0.984 0.983 0.979 0.972

6.109 11.423 15.772 19.028 8.871 14.303 18.566 34.479

3.269 3.899 4.579 4.447 3.784 5.276 5.115 6.87

RP isotherm parameters

Table 2. Isotherm Parameters Estimated by the Nonlinear Method of Analysis of Equilibrium Arsenic(III) Sorption Data on NHICO at pH ) 7.0 (( 0.2) and T ) 303 (( 1.6) K Langmuir isotherm parameters θ0 ( · 10 ) 2

Freundlich isotherm parameters Ka

KF ( · 102) n

R2

χ2

A

B

g

2.061 2.289 1.865 2.003 1.759 1.873 2.018 2.150

0.998 0.993 0.992 0.999 0.997 0.997 0.997 0.991

0.893 5.907 5.624 0.572 2.428 3.463 3.963 15.984

10.850 25.625 18.847 19.749 8.849 20.711 16.742 36.047

1.716 3.693 2.639 2.611 0.904 2.298 1.375 3.533

0.958 0.835 0.783 0.833 1.033 0.812 1.162 0.839

As(III) + ion added

R2

χ2 ( · 102)

mmol · kg1-

dm3 · mmol-1

R2

χ2 ( · 102)

(mmol · kg1-) · (dm3 · mg-1)1/n

PO43Ca2+ SiO32ClSO42Mg2+ no ion HCO3-

0.998 0.991 0.990 0.998 0.997 0.996 0.996 0.989

0.784 6.699 5.287 1.760 2.041 3.951 4.959 15.223

6.608 8.166 8.456 8.835 9.434 10.676 10.686 11.669

6.608 2.231 1.639 1.738 0.965 1.510 1.815 2.276

0.980 0.979 0.987 0.988 0.981 0.991 0.962 0.977

8.191 15.054 7.398 9.236 12.602 8.930 42.036 31.651

3.631 5.122 5.072 5.154 4.242 5.985 6.335 7.637

RP isotherm parameters

presence of background ions were laid between (8.0 and 10.1) kJ · mol-1 (Table S1 of SI), which indicated that the arsenic(III) sorption by NHICO took place with the ion-exchange/chemisorption41 phenomenon. Thermodynamic Parameters. Thermodynamic parameters for the arsenic(III) removal reaction with NHICO were estimated by using standard equations42 on the basis of the assumption that the activity coefficient of solutes added for working in solution was unity. The thermodynamic constant (Kc) was evaluated by using eq 7, taking the experimental data of equilibrium qe and Ce determined at T (( 1.6) ) (283, 293, 303, 313, and 323) K.

Kc ) qe /Ce

(7) -1

where qe/Ce is called the sorption affinity (dm · g ). The qe and Ce have their usual significance and are given elsewhere. The values of the changes of entropy and enthalpy at standard conditions (∆So and ∆Ho) were calculated from the slope and intercept of the plot of log(qe/Ce) versus (1/T) (Figure 3) of the linear relation shown (eq 8). 3

log10(qe /Ce) ) ∆S0 /2.303R - (∆H0 /2.303R)1/T o

o

(8)

o

Taking the values of ∆S and ∆H , the values of ∆G were calculated. The values of ∆Go estimated for the thermodynamic parameters are shown in Table 3. It was found that the reactions in the presence and absence of ions at the background were endothermic (∆Ho ) positive), which took place with increasing entropy (∆So ) positive) at the solid-liquid interface, and spontaneous (∆Go ) negative). The increase of entropy was due to the increase of randomness at the solid-liquid interface due to the release of aqua molecules when aquatic arsenic(III) species was distributed on the solid surface. The values of ∆Go were found to be increasingly negative with increasing temperature on the reactions in the absence and presence of background ions, indicating the increase in spontaneity of the reactions.

Figure 3. Plots of ln(qe/Ce) versus (T-1 · 10-3)/K-1 for the sorption reaction of arsenic(III) with NHICO from arsenic(III) solutions at CI (mmol · dm-3) ) (a) 0.067 and (b) 0.133 in the presence of some background ions. 9, no ion; b, SO42-; 2, Cl-; 1, PO43-; [, HCO3-; left-pointing triangle, Mg2+; right-pointing triangle, Ca2+; and `, SiO32-.

2044

Journal of Chemical & Engineering Data, Vol. 55, No. 5, 2010

Table 3. Thermodynamic Parameters Estimated for Arsenic(III) Sorption on NHICO at Different Reaction Temperatures (K) and at pHi 7.0 (( 0.2) As(III) + ion added no ion SO42ClPO43HCO3Mg2+ Ca2+ SiO32-

CI

∆H° -3

mmol · dm 0.067 0.134 0.067 0.134 0.067 0.134 0.067 0.134 0.067 0.134 0.067 0.134 0.067 0.134 0.067 0.1340

-∆G°/(kJ · mol-1)

∆S° -1

kJ · mol

30.314 27.353 15.952 19.264 12.679 25.117 24.775 20.058 38.271 32.721 17.647 27.461 15.191 13.354 20.169 19.688

-1

kJ · mol

-1

K

0.165 0.152 0.109 0.122 0.106 0.143 0.140 0.123 0.198 0.177 0.123 0.151 0.115 0.105 0.135 0.129

Mass Transfer Kinetics. The transfer of mass from the liquid to the solid phase carries out the uptake of a solute in the liquid phase by the surface of the sorbent, which may take place in a number of steps. McKay and Ho43 developed a model based on the assumptions of the (i) mass transfers from the aqueous phase onto the solid surface, (ii) adsorption of solute onto the surface sites, and (iii) internal diffusion of the solute via either a pore diffusion model or homogeneous solid-phase diffusion model. Taking into account these probable steps, the modified McKay and Ho43 model eq 9 was used for the present investigation.

ln[Ct /CI - 1/(1 + mK1)] ) ln[mK1 /(1 + mK1)] (1 + mK1)β1SSt

283 K

293 K

303 K

313 K

323 K

16.381 15.776 14.923 15.148 17.347 15.352 14.788 14.666 17.735 17.31 17.078 15.357 17.297 16.474 18.093 16.819

18.031 17.300 16.014 16.364 18.408 16.782 16.885 15.893 19.714 19.081 18.305 16.870 18.445 17.528 19.445 18.109

19.681 18.824 17.105 17.580 19.469 18.212 17.584 17.120 21.693 20.849 19.532 18.383 19.593 18.582 20.797 19.399

21.331 20.348 18.742 18.796 20.530 19.642 18.982 18.347 23.672 22.617 20.759 19.896 20.741 19.636 22.149 20.689

22.981 21.872 19.287 20.012 21.591 21.072 20.380 19.574 25.651 24.385 21.986 21.409 21.889 20.690 23.501 21.979

This is presumably due to the increased coulombic inhibition between the solute species in solution and on solid surface being developed by rapid uptake at initial stages from the solution of high CI value. Filtration of High Arsenic Groundwater Using a Fixed-Bed NHICO Column. Figure 5 demonstrates the result that is obtained by filtering the collected groundwater contami-

(9)

where CI and Ct are concentrations (mmol · dm-3) of arsenic at time t ) 0 and t ) equilibrium, respectively; m is the mass (g) of the adsorbent per unit volume, K1 is a constant which is the product of the Langmuir monolayer capacity (θ0) and Langmuir equilibrium constant (Ka). β1 is the mass transfer coefficient (m · s-1), and SS is the outer specific surface (m-1) of the adsorbate per unit volume. Taking the time-dependent residual arsenic(III) concentration (Ct, mmol · dm-3) obtained from the sorption kinetics experiment (Figure S8a,b of SI), the values of ln[Ct/CI - 1/(1 + mK1)] were calculated and plotted against time, t (min). Figure 4a,b shows the linear plots with slopes [{(1 + mK1)/mK1}β1SS]. The values of the correlation coefficient (R2) ranged in 0.95 to 0.99 indicated good linearity of the plots. The values of mass transfer coefficients (β1) calculated from the slopes are given in Table S2 of SI. It was found that the values of β1 were in general of the order 10-13. Only the prefix coefficients differed, indicating a slight effect on the mass transportation process. The values of β1 (Table S2 of SI) for the case of CI ) 0.067 mmol · dm-3 were found to be higher in the presence of Mg2+, Cl-, SiO32-, and HCO3- than in the presence of no or other background ions, while that was not clearly found when the CI of arsenic(III) was 0.172 mmol · dm-3; that is, the mass transportation of arsenic(III) from solution onto the solid surface was higher in the presence of Mg2+, Cl-, SiO32-, and HCO3-, especially when CI was less. Again, the value of β1 was higher at CI ) 0.067 mmol · dm-3 than CI ) 0.172 mmol · dm-3 of arsenic(III) solution when no background ion was added, indicating higher mass transportation from the aqueous solution on the solid surface at low concentrations.

Figure 4. Mass transfer kinetic plots of arsenic(III) sorption on NHICO from the solutions of CI (mmol · dm-3) ) (a) 0.067 and (b) 0.172 at T ) 303 (( 1.6) K and pHi ) 7.0 (( 0.2) in the presence of some background ions. 9, no ion; b, SO42-; 2, Cl-; 1, PO43-; [, HCO3-; left-pointing triangle, Mg2+; right-pointing triangle, Ca2+; and `, SiO32-.

Journal of Chemical & Engineering Data, Vol. 55, No. 5, 2010 2045

δ/D ) tδ /(tx - tf)

(12)

where tf is the time required for initial formation of the primary adsorption zone. The portion marked by ABC in Figure 5 represents the amount of solute or impurity adsorbed by NHICO in the PAZ from the break point to exhaustion. This quantity, MS, may be calculated by integrating the quantity (Ci - C) over Ve between the limits of VX and Vb. This does not represent the total capacity of NHICO within the PAZ at the break point. However, the total capacity would be given by the product, (VX - Vb) · Ci. Thus, one can define a fractional capacity, f, of the adsorber in the adsorption zone at break point to continue to remove solute from solution by eq 13.

∫ (Ci - C)·dVe/(VX - Vb)·Ci

f ) MS /(VX - Vb)·Ci )

(13) where Ci and C are input and effluent concentrations of solute in terms of the mass passing a unit cross-sectional area of the adsorber. The fractional capacity, f, approaches zero as the time required for initial PAZ formation, tf, approaches the time, tδ, required for the zone to move a distance in the bed equal to its own depth, δ. For systems in which the formation time, tf, is very small (approaching zero), the value of f approaches unity.

Figure 5. Breakthrough curve for the filtration of a groundwater sample by a NHICO packed fixed-bed column.

nated with high arsenic (1.347 µmol · dm-3). To estimate the rate of attainment of equilibrium of arsenic concentration between mobile and stationary phases, the breakthrough curve (Figure 5) that is obtained by the treatment of field sample was analyzed.44 The curve was idealized by the assumption that the removal of the solute is complete over the initial stages of operation. The break point was chosen at a point of arsenic concentration (Cb) crossed over 0.01 mg · dm-3 (0.133 µmol · dm-3) in the effluent. At an arbitrarily selected effluent concentration, Cx, closely approaching CI, the sorbent was considered to be essentially exhausted. The primary adsorption zone (PAZ) in the fixed-bed adsorber represented by the breakthrough curve in Figure 5 is defined as that part of the bed over which there is a concentration reduction from Cx to Cb. It is assumed that this zone is of constant length or depth, δ. The total time, tx, involved for the primary zone to establish itself and move down the length of the adsorber and out of the bed may be calculated by eq 10.

tx ) Vx /Fm

f ) 1 - tf /tδ

Thus, by combining eqs 13 and 14, the following eq 15 results,

δ/D ) tδ /[tx + tδ(f - 1)]

(15)

The percent saturation at this point is given by the eq 16

% saturation ) [D - δ(f)]/D·100

(16)

The experimental and calculated parameters for the fixedbed sorption of arsenic by the NHICO bed were summarized respectively in Table 4a,b. It was found that the fixed-bed column of NHICO filtered 4.7 dm3 (997.9 BV) of water with arsenic concentration below 0.01 mg · dm-3 (0.133 µmol · dm-3). The column saturation at exhaustion point is estimated to be 81.5 %. The used bed was regenerated up to level of 90 (( 1.5) % of its initial activity with 0.350 dm3 of optimized 0.25 M NaOH solution.

(10)

Fm is the mass rate of flow to the adsorber, expressed as mass per unit time per unit cross-sectional area of the bed. The time, tδ, required for movement of the zone down its own length in the column after it has been established is simply (eq 11)

tδ ) (Vx - Vb)/Fm

(14)

Conclusion Synthetic hydrous iron(III) + chromium(III) mixed oxide (NHICO), agglomerated nanoparticles (40 to 50 nm), heattreated at 623 K showed the highest sorption affinity for arsenic(III) from aqueous solutions at pH ) 7.0 (( 0.2) and T ) 303 (( 1.6) K. Use of this material for estimating

(11)

Thus, for a depth D of the adsorber, one may equate the depth and time ratios (eq 12): Table 4

a. Observed Parameters for the Fixed-Ced Arsenic(III) Removal by NHICO Column flow rate ( · 10-2) -1

Co ( · 10-5)

Cx ( · 10-5)

-3

-3

dm · h

mmol · dm

mmol · dm

6.0

134.667

127.701

3

Cb ( · 10-5) -3

mmol · dm 8.108

Vx ( · 104) -2

kg · m

11.904

Vb ( · 104) -2

kg · m

9.350

(Vx - Vb) ( · 104) -2

Fm ( · 103) -1

kg · m

D ( · 10-2)

-2

kg · h · m

2.553

m

1.558

6.0

b. Calculated Parameters for the Fixed-Bed Arsenic(III) Removal by NHICO Column flow rate ( · 10-2)

tx

tδ

dm3 · h-1

h

h

6.0

76.381

16.383

tf

δ ( · 10-2)

f

h

m

% saturation

0.831

2.768

1.335

81.5

2046

Journal of Chemical & Engineering Data, Vol. 55, No. 5, 2010

equilibrium and thermodynamics arsenic(III) sorption from aqueous solutions in the presence of different background ions that occur in groundwater showed that the reactions took place obeying Langmuir and RP isotherm models. The estimated Langmuir monolayer capacity (θ0; 10.636 · 102 mmol · kg1-) was higher than that obtained in the presence of investigated background ions except for HCO3-, which showed notable positive effects on the equilibrium. Whatever the background ion is, the changes of enthalpy (∆H0) and entropy (∆S0) were positive, indicating an endothermic arsenic(III) sorption reaction driven by entropy increase at the solid-liquid interface. The negative free energy (∆G0) changes indicated that the reactions were spontaneous and increased with increasing temperature on the reaction. The NHICO packed column (height: 6.0 cm, internal diameter: 1.0 cm, BV: 4.71 cm3), when used for filtration of high arsenic groundwater (As concentration: 1.347 µmol · dm-3), yielded 4.7 dm3 (∼ 1000 BV) of water with arsenic concentration of e 0.133 µmol · dm-3. The bed after use was regenerated up to a level of 90 (( 1.5) % with 0.350 dm3 of optimized 0.25 M NaOH solution. Acknowledgment We are thankful to the Head, Department of Chemistry and the Principal, Presidency College, Kolkata, India for providing laboratory facilities. Supporting Information Available: Some details of the characterization of NHICO, heat treatment, and pH effects with some figures and tables. This material is available free of charge via the Internet at http://pubs.acs.org.

Literature Cited (1) Meng, X. G.; Jing, C.; Korfiatis, G. P. A Review Redox Transformation of Arsenic in Environments. ACS Symp. Ser. 2003, 835, 70–83. (2) Smedley, P. L.; Kinniburgh, G. A Review of the Source, Behavior and Distribution of Arsenic in Natural Waters. Appl. Geochem. 2002, 17, 517–568. (3) McArthur, J.; Ravenscroft, P.; Safiullah, S.; Thirwall, M. F. Arsenic in Groundwater: Testing Pollution Mechanisms for Sedimentary Aquifers in Bangladesh. Water Res. 2001, 37, 109–117. (4) Harvey, C. F.; Swartz, C. H.; Baduzzaman, M.; Keon-Blute, A. B.; Yu, W.; Ali, M. A.; Ray, J.; Beckie, R.; Niedon, V.; Brabander, D.; Oates, P. M.; Asfaque, K. N.; Islam, S.; Hemond, H. F.; Ahmed, M. F. Arsenic Mobility and Groundwater Extraction in Bangladesh. Science 2002, 298, 1602–1606. (5) Mondal, P.; Majumder, C. B.; Mohanty, B. Laboratory Based Approaches for Arsenic Remediation from Contaminated Water: Recent Developments. J. Hazard. Mater. 2006, B137, 464–479. (6) Mohan, D.; Pittman, C. U., Jr. Arsenic Removal from Water/ Wastewater using Adsorbents- a Critical Review. J. Hazard. Mater. 2007, 147, 1–53. (7) Reed, B. E.; Vaughan, R.; Jiang, L. As(III), As(V), Hg and Pb Removal by Fe-oxide Impregnated Activated Carbon. J. EnViron. Eng. 2000, 126, 869–873. (8) Kundu, S.; Gupta, A. K. Investigation on the Adsorption Efficiency of Iron Oxide Coated Cement (IOCC) towards As(V) - Kinetics, Equilibrium and Thermodynamics Studies. Colloid Surf., A 2006, 273, 121–128. (9) Kundu, S.; Gupta, A. K. Adsorption Characteristic of As(III) from Aqueous Solution on Iron Oxide Coated Cement (IOCC). J. Hazard. Mater. 2007, 142, 97–104. (10) Wilkie, J. A.; Hering, J. G. Adsorption of Arsenic onto Hydrous Ferric Oxide: Effects of Adsorbate/Adsorbent Ratios and Co-occurring Solutes. Colloids Surf., A 1996, 107, 97–110. (11) Driehaus, W.; Jekel, M.; Hilderbrandt, U. Grannular Ferric Hydroxidea New Adsorbent for the Removal of Arsenic from Natural Water. J. Water Supply: Res. Technol.sAQUA 1998, 47, 30–35. (12) Raven, K. P.; Jain, A.; Loeppert, R. H. Arsenite and Arsenate Adsorption on Ferrihydrite: Kinetics, Equilibrium, and Adsorption Envelopes. EnViron. Sci. Technol. 1998, 32, 344–349.

(13) Altundogan, H. S.; Altundogan, S.; Tumen, F.; Bildik, M. Arsenic Adsorption from Aqueous Solution by Activated Red Mud. Waste Manage. 2002, 22, 357–363. (14) Singh, T. S.; Pant, K. K. Equilibrium, Kinetics and Thermodynamics Studies for Adsorption of As(III) on Activated Alumina. Sep. Purif. Technol. 2004, 36, 139–147. (15) Lin, T. F.; Wu, J. K. Adsorption of Arsenite and Arsenate within Activated Alumina Grains: Equilibrium and Kinetics. Water Res. 2001, 35, 2049–2057. (16) Katsoyiannis, I. A.; Zouboulis, A. I. Removal of Arsenic from Contaminated Water Sources by Sorptoin onto Iron Oxide-Coated Polymeric Materials. Water Res. 2002, 36, 5141–5155. (17) Thirunavukkarasu, O. S.; Viraraghavan, T.; Subramanian, K. S. Arsenic Removal from Drinking Water using Iron Oxide Coated Sand. Water, Air, Soil Pollut. 2003, 142, 95–111. (18) Zeng, L. Arsenic Adsorption from Aqueous Solution on an Fe(III)-Si Binary Oxide Adsorbent. Water Qual. Res. J. Can. 2004, 39, 269– 277. (19) Kuriakose, S.; Singh, T. S.; Pant, K. K. Adsorption of As(III) from Aqueous Solution onto Iron Oxide Impregnated Activated Alumina. Water Qual. Res. J. Can. 2004, 39, 258–268. (20) Hlavay, J.; Polyak, K. Determination of Surface Properties of Iron Hydroxide-Coated Alumina Adsorbent for Removal of Arsenic from Drinking Water. J. Colloid Interface Sci. 2005, 284, 71–77. (21) Lenoble, V.; Bouras, O.; Deluchat, V.; Serpaud, B.; Bollinger, J. C. Arsenic Adsorption onto Pillared Clays and Iron Oxides. J. Colloid Interface Sci. 2002, 255, 52–58. (22) Manna, B. R.; Dey, S.; Debnath, S.; Ghosh, U. C. Removal of Arsenic from Ground Water using Crystalline Hydrous Ferric Oxide (CHFO). Water Qual. Res. J. Canada 2003, 38, 193–210. (23) Manna, B. R.; Dasgupta, M.; Ghosh, U. C. Studies on Crystalline Hydrous Titanium Oxide (CHTO)- as Scavenger of Arsenic (III) from Natural Water. J. Water Supply: Res. Technol.sAQUA 2004, 53, 483– 495. (24) Manna, B. R.; Debnath, S.; Hossain, J.; Ghosh, U. C. Trace Arseniccontaminated Groundwater Upgradation using Hydrated Zirconium Oxide (HZO). J. Ind. Pollut. Control 2004, 20, 247–266. (25) Manna, B. R.; Ghosh, U. C. Adsorption of Arsenic from Aqueous Solution by Synthetic Hydrous Stannic Oxide. J. Hazard. Mater. 2007, 144, 522–531. (26) Ghosh, U. C.; Bandhyopadhyay, D.; Manna, B. R.; Mandal, M. Hydrous Iron(III)-Tin(IV) Binary Mixed Oxide: Arsenic Adsorption Behaviour. Water Qual. Res. J. Can. 2006, 41, 198–209. (27) Manna, B. R.; Ghosh, U. C. Pilot-Scale Performance of Arsenic and Iron Removal from Contaminated Groundwater. Water Qual. Res. J. Can. 2005, 40, 82–90. (28) Zhang, W. X. Nano-scale Iron Particles for Environmental Remediation: an Overview. J. Nanopart. Res. 2003, 5, 323–332. (29) Kanel, S. R.; Manning, B.; Charlet, L.; Choi, H. Removal of Arsenic(III) from Groundwater by Nanoscale Zero-valent Iron. EnViron. Sci. Technol. 2005, 39, 1290–1298. (30) Yuan, C.; Lien, H.-L. Removal of Arsenate from Aqueous Solution using Nanoscale Iron Particles. Water Qual. Res. J. Can. 2006, 41, 210–215. (31) Pena, M. E.; Korfiatis, G. P.; Patel, M.; Lippincott, L.; Meng, X. Adsorption of As(V) and As(III) by Nanocrystalline Titanium dioxide. Water Res. 2005, 39, 2327–2337. (32) Jing, C.; Meng, X.; Liu, S.; Baidas, S.; Patraju, R.; Christodoulatos, C.; Korfiatif, G. P. Surface Complexation of Organic Arsenic on Nanocrystalline Titanium dioxide. J. Colloid Interface Sci. 2005, 290, 14–21. (33) Hossain, M. A.; Sengupta, M. K.; Ahamed, S.; Rahaman, M. H.; Mondal, D.; Lodh, D.; Das, B.; Nayak, B.; Roy, B. K.; Mukherjee, A.; Chakraborty, D. Ineffectiveness and poor reliability of arsenic removal plants in West Bengal, India. EnViron. Sci. Technol. 2005, 39, 4300–4306. (34) Standard Methods for the Examination of Water and Wastewater, 20th ed.; Clesceri, L. S., Greenberg, A. E., Eaton, A. D., Eds.; APHA, AWWA, WEF: Washington, DC, 1998; pp 4-82. (35) Faust, S. D.; Aly, O. M. Adsorption Processes for Water Treatment; Butterworths: London, 1987. (36) Zhang, F.-S.; Itoh, H. Iron Oxide-loaded Slag for Arsenic Removal from Aqueous System. Chemosphere 2005, 60, 319–325. (37) Villalobos, M.; Leckie, I. O. Carbonate Adsorption on goethite under closed and open CO2 systems. Geochim. Cosmochim. Acta 2000, 64, 3787–3802. (38) Stachowicz, M.; Hiemstra, T.; Van Riemsdijk, W. H. Surface Speciation of As(III) and As(V) Adsorption in Relation to Charge Distribution. J. Colloid Interface Sci. 2006, 302, 62–75. (39) Hall, K. R.; Eagleton, A. A.; Vermeulen, T. Pore- and solid diffusion kinetics in fixed-bed adsorption under constant-pattern conditions. Ind. Eng. Chem. Fundam. 1966, 5, 212–223.

Journal of Chemical & Engineering Data, Vol. 55, No. 5, 2010 2047 (40) Dubinin, M. M.; Radushkevich, L. V. The equation of the characteristic curve of activated charcoal. Proc. Acad. Sci. USSR Phys. Chem. Sect. 1947, 55, 331. (41) Reiman, W.; Walton, H. Ion-Exchange in Analytical Chemistry, International Series of Monographs in Analytical Chemistry, Vol. 38; Pergamon Press: Oxford, 1970. (42) Atkins, P.; de Paula, J. Atkin’s Physical Chemistry, 8th ed.; Oxford University Press: Oxford, 2006. (43) Ho, Y. S.; McKay, G. A. Comparison of Chemisorption Kinetics Models Applied to Pollutant Removal on Various Sorbents. Trans Inst. Chem. Eng. 1998, 76 (Part B), 332–340.

(44) Waber, J., Jr. Physicochemical Processes for Water Quality Control; John Wiley & Sons, Inc.: New York, 1972.

Received for review November 28, 2009. Accepted March 5, 2010. Authors are grateful to University Grants Commission, New Delhi (India) for financial support of the project (No. F. 32-272/2006 (SR)).

JE901010X